Using NHET pins as simple inputs

Hi Michael

You have to read DIN register.

I am reading the DIN register.

I'm trying to read NHET1.7 and NHET1.9.  This my code to read the pins...

    uint8 result = 0;

    uint32 hetDin = hetPORT1->DIN;

    if ((hetDin & 0x0020) == 0x0020) {

        result |= 0x01;

    }

    if ((hetDin & 0x0008) == 0x0008) {

        result |= 0x02;

    }

I'm not just trying to read any old HET pins.  I'm trying to read the digital hall sensor on the DRV8301-LS12 kit.  If I've misinterpreted the schematic and I'm looking at the wrong pins, please let me know.

Thanks,

Mike

Hi Anthony,

I can spin the motor by hand and see the halls on my oscilloscope.  I get the pattern I expect.

I setup the debugger on refresh and looked at the HET->DIN register.  I see these bit toggling as I turn the motor: 0x1000000A.  That is bits 28, 3, & 1.  None of these pins' state correlates to the signals I see on the scope.  And my expectation is to see pins 7 & 9 toggling.

The hall sensor that is connected to GIOB5 works fine (what I see with the scope relates to what I see in the debugger and sw).

Thanks,

Mike

NHET7 and NHET9 are connected through 50 ohm resistors to ECAP6 and ECAP5 respectively.  I assumed that ECAP 5 & 6 were inputs so I didn't pay much attention to them, but by default the pins are not muxed to ECAP 5 & 6.

When I change my HALCoGen project to mux the pins to ECAP 5 & 6, rebuild and retest, it does not make a difference.

Mike,

Huh - this is turning into quite a puzzle.

Pins N2HET1[3] and N2HET1[1] are multiplexed with the EQEP2 module,  which goes to pins 40 and 90 on the DIMM.

Could your hall sensor hookup be going to theset pins?   Which connector on the DRV8301 board did you connect the sensor to?

Thanks and Best Regards,

Anthony

The Teknic motor comes with two 5 pin headers, one is the Encoder interface, which should be connected to J4 in the DRV8301 monther board, the other one is the Hall sensor interface, which should be connected to the J10 in the DRV8301. J4 and J10 are next to each other close to the edge of the board.

Sounds to me that you connect the hall sensor to J4 instead of J10.

I attached part of the code to run with the Hall sensor with this kit. The total project is too big to be posted. The sys_main.c starts with the pin mux code (I did not use the HalCoGen pin mux). rm46hall_gpio_BLDC.c is the code to get the hall sensor information. 7612.sys_main.c2086.rm46hall_gpio_BLDC.c0647.rm46hall_gpio_BLDC.h

 hetREG1->DIR = hetREG1->DIR & 0xDFFFFD7F;//(1<<29)+(1<<7)+(1<<9)
 hetREG1->SFPRLD = 5; //5 VLCK Counter delay amount to validate/debounce GPIO readings
 hetREG1->SFENA = 0x20000280;
 temp = hetREG1->DIN;
 hall->HallGpio =  ((temp >>29) & 0x1) +((temp >>6) & 0x2) + ((temp >>7) & 0x4);

Good Luck!

Haixiao

The cables from the motor are label J4 and J10 and I have them plugged into to the right connectors on the board.

Michael,

I checked the 8301 board document and they said J10 is connected to CAP1/2/3 hall sensor 1/2/3 respectively.

The CAP 1/2/3 are connected to DIM100 pin 30,80,31 respectively.

That translate to HET1[13],HET1[17], GIOB5 respectively.

can you please double the connection again?

Michael,

Look at the attached document page 11 (table 4-18) and page 8.

1222.DRV830x-HC-C2-KIT_HWGuide.pdf

Mike,

I think the idea that the cables are swapped matches what you are reporting, so it's highly likely that this is the issue.

Look at the picture below, you can see how swapping the cables would move the signals from HET1[7],[9] to HET1[3],[1]....

It could be a quality issue with the cable, maybe the labels are on the wrong cables? 

Probably the only way to prove this is to ohm out a few connections from the J4/J10 header to the motor side connector.

I believe this is the motor that comes with the kit, there's a pinout for the motor connector in this document:

http://www.teknic.com/files/product_info/N23_Industrial_Grade_Motors_v5.2.pdf

Mike,

Sorry - one more thought.

It may be that the cables *are* labeled correctly, and you've got them going to the correct headers on the DRV8301 board;  but have the J10 connector 'backward'.

This would also result in you seeing the encoder inputs toggle but no activity on the hall sensor as you rotate the rotor.

In the quick start guide there is a color photo showing the wire color order.  You might also check that the wire color order there matches how you have the cables plugged into the board.   There isn't any sort of key - so it would be very easy to flip this around.

Best Regards,

-Anthony 

Excellent work Mike.

Looks like the 4K resistor in the DRV board are actually in series with the output on the level shifter, so no matter how strong you would drive the signal on J10, you would have this problem. 

Thanks for debugging this one.   Looks like we have something to fix.

Best Regards,
Anthony